ICAE International Commission on Atmospheric Electricity

ICAE 2003 Versailles

ICAE Home Page

General Information

Accommodation /
Accompanying Person

Benjamin Franklin

Conference Format

Notice to Chairpersons


Index to Authors




Thursday 11th June



Session C1 Physics of Lightning I

8:30 N L Aleksandrov, E M Bazelyan and Y P Raizer
keynote: Initiation and Development of Lightning Discharge: Physical Mechanism and Problems

N. S. Khaerdinov, A. S. Lidvansky, and V. B. Petkov
Effect of the Electric Field of thunderclouds on Cosmic Rays and evidence for pre-lightning acceleration of electrons

9:15 H. E. Tierney, R. A. Roussel-Dupré, E. M. D. Symbalisty, and L. Triplett
Runaway Breakdown and Lightning Initiation
9:30 A. Larsson, A. Delannoy and P. Lalande
The voltage gradient along a lightning channel during strikes to aircraft
9:45 J-P. Pinty, G. Molinié, C. Barthe, and F. Roux
A semi-deterministic scheme to produce IC/CG lightnings in a 3D cloud resolving model
10:00 W. Rison, P. Krehbiel, R. Thomas, T. Hamlin, and J. Harlin
Lightning Mapping and Radar Observations of Bolts from the Blue  
10:15 J. C. Willett, Garrett Park, D. M. Le Vine
Lightning Return-Stroke Current Waveforms Aloft, From Measured Field Change, Current, and Channel Geometry


Initiation and Development of Lightning Discharge:
Physical Mechanism and Problems

N.L. Aleksandrov
Moscow Institute of Physics & Technology, Dolgoprudny, Moscow region, Russia

E.M. Bazelyan
Krzhizhanovskyy Power Engineering Institute, Moscow, Russia

Y. P. Raizer
Institute for Problems in Mechanics, Russian Academy of Science, Moscow, Russia

This paper considers three interconnected problems: (i) initiation of a downward lightning in a thundercloud and that of an upward lightning near tall grounded structures; (ii) conditions for lightning development in the cloud-to-ground gap and the effects of lightning trajectory and branching on discharge parameters; and (iii) physical mechanism of the lightning return stroke and peculiarities of the propagation of current and voltage waves along the plasma channel with non-linear parameters. The focus is on the estimation of lightning discharge parameters required to solve applied problems including the simulation of lightning electromagnetic fields and frequency of lightning strokes to grounded and flying objects. Key problems of lightning physics which hinder the progress in practical lightning protection are posed.
TopFull program for C1 Session

Effect of the Electric Field of thunderclouds on Cosmic Rays
and Evidence for pre-lightning acceleration of electrons

N. S. Khaerdinov, A. S. Lidvansky, and V. B. Petkov
Institute for Nuclear Research, Russian Academy of Sciences, 60th October Anniversary pr.,7a, Moscow, 117312 Russia
Email: lidvansk@sci.lebedev.ru


The flux of secondary cosmic rays deep in the atmosphere is subject to variations that can be related to meteorological parameters. The disturbed electric field in thunderstorm and rainy periods is among them. We have studied the electric field effect recording separately soft (electrons) and hard (muons) components of secondary cosmic rays and correlating them with the near-earth electric field strength. The data were obtained with the Carpet air shower array at the Baksan valley (North Caucasus, 1700 m a.s.l.) in the summer seasons of 2000 and 2001. It is shown that, generally, the behavior of soft and hard components is reasonably well understood, being described by the so-called electron and muon mechanisms of electric field influence on cosmic ray particle fluxes. These mechanisms were discussed in literature long ago, but their manifestation is observed experimentally for the first time. As was predicted, in the electric field the net effect for the soft component is increased intensity, while the hard component mainly demonstrates the decrease of its intensity. The correlation with the field strength is linear and quadratic for the soft and hard components, respectively. In addition, some irregular enhancements of the soft component just before lightning strokes are observed, which look very similar to a particle acceleration process. One can consider this as a first direct confirmation of the runaway breakdown process, which is extensively discussed by theoreticians as a mechanism for lightning initiation. We believe that such experiments with cosmic rays can be a useful tool for studying the atmospheric electricity phenomena.


TopFull program for C1 Session

Runaway Breakdown and Lightning Initiation

Heidi E. Tierney, Robert A. Roussel-Dupré, Eugene M.D. Symbalisty, and Laurie Triplett
Atmospheric and Climate Sciences, Los Alamos National Laboratory, Los Alamos, New Mexico


Runaway breakdown is considered to be a viable mechanism for lightning initiation. Background levels of ~1-MeV cosmic ray electrons can avalanche in electric fields that are a factor of ten lower than those required for conventional breakdown. If a sufficient ambient electric field exists over a large spatial scale, such as in a thunderstorm, the high-energy electrons of the avalanche produce measurable levels of x-ray photons via the Bremsstrahlung process. Modeled x-ray energy spectra and count rates are in agreement with observations of x-ray bursts observed during thunderstorms. This success has motivated the investigation of radio and optical emissions from a lightning discharge involving high-energy electrons. A model of a runaway discharge, initiated by a cosmic ray secondary electron, in a thunderstorm electric field is presented. The dominant rates that control the evolution of the runaway discharge in the troposphere and in an external electric field are: avalanche ionization, low-energy electron attachment, and high-energy electron loss. The latest estimates (from theory and experiment) for the runaway avalanche rates are used. The self-consistent evolution of the electromagnetic fields is also included. The X ray, optical, and radio emissions from the primary and secondary electron currents are presented and compared with satellite and ground-based observations.


TopFull program for C1 Session

The voltage gradient along a lightning channel during strikes to aircraft

Anders Larsson
FOI - Swedish Defence Research Agency, Grindsjön Research Centre, SE-147 25 TUMBA, Sweden

Alain Delannoy, Philippe Lalande,
ONERA - DMPH, BP 72 - 29, avenue de la Division Leclerc, F-92322 CHATILLON CEDEX, France
Alain.Delannoy@onera.fr; Philippe.Lalande@onera.fr


Statistical in flight analysis on airliner shows that an aircraft is struck by lightning once per year. This threat is taken into account in the aircraft protection design and in certification processes through the concept of zoning. The aim of the zoning is to locate and classify surfaces on an aircraft as a function of the lightning threat expressed in the different components of the lightning current. In order to compute the zoning, Onera has developed two physical models, which simulate different phases of a lightning strike to an aircraft in flight. The first phase is short (5 ms) in regard with the aircraft motion and is associated with the initiation and development, from the aircraft, of electrical discharges which create the conductive channel in which the lightning current flows. The second phase lasts from 200 ms to 1 s and is associated with the flow of the lightning current. During this time, the aircraft motion deforms the channel and causes the arc root to sweep along the fuselage.

The modeling of this last phase depends on the arc channel characteristics. The purpose of the article is to present, from experimental and theoretical analysis, a relation between the current flowing the channel and the voltage gradient along the channel, which mainly drives the sweeping processes of the arc root.

The voltage gradient, E, along the channel has one resistive and one inductive part and is relied to the lightning current and channel properties according to the following expression

where I(t) is the lightning current, R is resistance per unit length of the channel and L is the inductance per unit length. During the continuing current phase, the resistive part dominates and during the return strokes and recoil leaders the inductive part dominates. The lightning current is an input data and can be derived from in-flight measurement campaigns or regulation document. The inductance of the channel is roughly constant about 2,0 ± 0,5 µH/m. For the resistive part, no expression was available. Thus, by combining various experimental and theoretical work, we have derived the following simple analytical expression between the voltage gradient along the channel and the continuing current :

where is a correction term for atmospheric conditions.


TopFull program for C1 Session

A semi-deterministic scheme to produce IC/CG lightnings in a 3D cloud resolving model

Jean-Pierre Pinty
Laboratoire d'Aérologie, CNRS, Obseravtoire Midi-Pyrénées, 14 avenue E.Belin Toulouse, F-31400, France

Gilles Molinié
NIS-1 Group Office (Space and Atmospheric Sciences), TA-3/1888/01U Bikini Atoll Rd., SM-30 Los Alamos National Laboratory P. O. Box 1663, MS D466 Los Alamos, NM 87545 USA
505 667 6104

Christèle Barthe
Laboratoire d'Aérologie, CNRS, Observatoire Midi-Pyrénées, 14, avenue E. Belin, Toulouse, F-31400 France

Frank Roux
Laboratoire d'Aérologie, CNRS, Observatoire Midi-Pyrénées, 14, avenue E. Belin, Toulouse, F-31400 France


The explicit simulation of electric charging processes is performed in a 3D cloud resolving model. The electric field, E, is solved by integrating an elliptical equation for the electric potential at each time step. The growth of E is bounded below a triggering level by an efficient lightning scheme which operates in two stages. First, once E exceeds a threshold value a lightning is initiated and two leaders propagate in opposite directions of maximum E, until E falls below a stopping value. Second, a branching algorithm is set to account for the tortuous aspect of the flash propagation in the manner of dielectric breakdown models. The iterative algorithm seeks for forking gridpoints on the bi-leader channel that fulfill charge and potential requirements. In order, to grossly reproduce the fractal dimension of the simulated lightning, a stochastic stepwise branching is generated with the number of branches or streamers prescribed by a scaling law. The full scheme will be described in the extended abstract and some important properties will be outlined. Owing to its high computational efficiency (parallelized coding), this complex scheme is currently used to simulate the electrical activity of 3D storms.


TopFull program for C1 Session

Lightning Mapping and Radar Observations of Bolts from the Blue

William Rison, Paul Krehbiel, Ron Thomas, Tim Hamlin and Jeremiah Harlin
Langmuir Laboratory for Atmospheric Research New Mexico Institute of Mining and Technology Socorro, New Mexico 87801


A "bolt from the blue" is a cloud-to-ground lightning discharge which exits from the side of a thundercloud and comes to ground away from the thundercloud. In several field projects we have simultaneous observations of negative "bolt from the blue" lightning discharges from the New Mexico Tech Lightning Mapping Array (LMA) and dual-polarization radars. These flashes begin as normal polarity intracloud discharges, with the initial breakdown in the strong field region between the main negative charge and the upper positive charge. Negative streamers propagate from the initial breakdown region upward into the upper positive charge, and spread horizontally through the upper positive charge. After neutralizing much of the upper positive charge, an excess of negative charge remains in the cloud, and it is energetically favorable for the discharge to continue toward the positive image charge below the ground. The discharge progresses to the edge of the cloud, and often follows a screening layer along the cloud boundary for a while. The discharge then develops downward away from the cloud as a stepped leader. The ground strike point can be as far as several tens of kilometers horizontally from the cloud boundary.

The reason a thunderstorm produces bolts from the blue is apparently related to the charge structure of the storm. The main negative charge is considerably larger than the upper positive charge, so the upper positive charge cannot neutralize all the negative charge in the cell. Also the lower positive charge is relatively weak, so that the electric field in the region between the main negative and lower positive is not strong enough to initiate a breakdown.


TopFull program for C1 Session

Lightning Return-Stroke Current Waveforms Aloft, From Measured Field Change, Current, and Channel Geometry

J.C. Willett,
P.O. Box 41, Garrett Park, MD 20896, USA; D.M. Le Vine, Code 975, NASA/GSFC, Greenbelt, MD 20771, USA


Direct current measurements are available near the attachment point from both natural cloud-to-ground lightning and rocket-triggered lightning, but relatively little is known about the return-stroke currents aloft. We present, as functions of height, current amplitudes, rise times, and effective propagation speeds that have been estimated with a novel technique [Willett and Le Vine, Proceedings, 10th International Conference on Atmospheric Electricity, Osaka, Japan, 10-14 June 1996] from data on 24 subsequent return strokes in six different lightning flashes that were triggering at the NASA Kennedy Space Center, FL, during 1987. The unique feature of this data set is the stereo pairs of still photographs, from which three-dimensional channel geometries were determined previously [Willett and Le Vine, AGU Fall Meeting, San Francisco, CA, December, 1995]. Having channel geometry permits us to calculate the fine structure of the electric-field-change waveform, E(t), produced by each stroke, using the corresponding current waveform measured at the channel base [Leteinturier and Hamelin, IEEE Trans. EMC, 33, 351-357, 1991] together with physically reasonable assumptions about the current distributions aloft. The computed waveforms are compared with observed E(t) from the same strokes [Willett et al., J. Geophys. Res., 94, 13,275-13,286, 1989], and our assumptions are adjusted to maximize agreement.

An example of these calculations is given in the figure below. The black curve is measured E(t) for the third stroke in flash 8725. Compare first with the green curve, which is produced by propagating the measured current waveform up the reconstructed 3-D channel, without change of shape or amplitude, at a constant speed of 1.9 x 108 m/s. The amplitude of this computed waveform is too large, and it has too much high-frequency structure; but its major (low-frequency) features have been made to coincide with the observed fine structure by using a propagation speed 12% larger than the two-dimensional velocity that was measured by a streak camera. Smoothing the observed current waveform (by convolution with a simple kernel, whose equivalent width increases with height from zero at the ground to 2.2 56;s at the top of the 1842 m long reconstructed channel) eliminates the excessive high-frequency structure, and decreasing the current amplitude by 45% with a 20 m exponential height scale corrects the E(t) amplitude. The result of these two adjustments is the red curve, in good agreement with observation.


In spite of the non-uniqueness of solutions derived by this technique, several conclusions seem inescapable:

1) The effective propagation speed of the current up the channel is usually appreciably (28 ± 13 %) faster than the 2-D velocity that was measured for 14 of these strokes.
2) Given the deduced propagation speed, the peak amplitude of the current waveform often must decrease dramatically with height (by 36 ± 6 % with a scale height of only 10 - 100 m) to prevent the amplitude of E(t) from being over-predicted.
3) The 10 - 90 % rise time of the current wave front must always increase rapidly with height (from the observed 325 ± 110 ns at the surface to 964 ± 95 ns at 300 m) in order to keep the fine structure of the calculated field consistent with the observations. [Means and standard deviations in (2) and (3) above apply to six events -- the largest stroke in each flash.]


TopFull program for C1 Session

Last Update : June 3, 2003
[ ICAE 2003 Home Page] - [ ICAE Contacts ]